
Huntsville, Alabama has a business mix few cities share: NASA's Marshall Space Flight Center and several Army commands on Redstone Arsenal, hundreds of aerospace, defense and IT firms in and around Cummings Research Park, and a long tail of small subcontractors doing the machining, testing, software and network work behind them. Most of those small firms grow on reputation and referrals, and more and more on being found by an engineer or a buyer who searches before calling, which is why capability pages and SEO services in Huntsville get as much of a founder's attention as the office lease. Far fewer of them have a written plan for the week the building, the network or the prime contract goes away. This piece is for them.
A ten-person cybersecurity or engineering services company does not need a binder the size of a bank's. It needs honest answers to five questions that come with working in the Tennessee Valley for federal customers. Below are those questions, with what we have seen work.
Weather: plan for the week, not the afternoon
North Alabama sits in the part of the Southeast that forecasters call Dixie Alley. The National Weather Service office in Huntsville keeps a tornado climatology for the region that shows the main season in spring and a second, smaller one in late fall and early winter. Tornadoes here often arrive at night, wrapped in rain, and move fast.
The day that shaped local planning was 27 April 2011, when a long outbreak of tornadoes crossed the state. Storms brought down major TVA transmission lines, the reactors at Browns Ferry shut down automatically, and much of north Alabama was without power for days. Plenty of offices in the research park came through without a scratch and still sat dark all week.
Lessons for a small firm:
- Assume a multi-day utility outage, not a two-hour one. A UPS buys time to shut servers down cleanly; it does not run an office. Decide in advance whether you will run a generator, move work to the homes that still have power, or stop and tell customers when you expect to restart.
- Know where your people live. In 2011 the damage was patchy. Some employees had power and internet, others had no street left. A short list of home towns (Madison, Decatur, Athens, Owens Cross Roads, the Tennessee line) tells you who can carry work on day two.
- Agree on a check-in rule. After a night warning, everyone posts in one channel or texts one person by 9 a.m. A named person owns the list of who has not answered and starts calling at 9:15.
- Move the phone number. Many small firms still ring a desk phone system in the office. Make sure forwarding to mobiles can be switched on from outside the building, by more than one person.
Controlled information changes the recovery plan
Many corridor firms hold controlled unclassified information (CUI) for a defense customer. That brings rules a generic continuity template ignores. DFARS clause 252.204-7012 requires a contractor to report a cyber incident that affects covered defense information within 72 hours of discovery, and to preserve images of the affected systems for at least 90 days. The Cybersecurity Maturity Model Certification program, described on the Defense Department's CMMC overview page, started showing up in solicitations after its acquisition rule took effect in November 2025, and its Level 2 requirements follow the 110 security requirements of NIST SP 800-171.
For continuity this means three things:
- Backups of CUI must live somewhere that meets the same rules. A consumer cloud sync folder is not a recovery site for controlled data. Some firms use a government cloud tenant, others keep an encrypted copy in a second secured location. Whichever you choose, the system security plan and the continuity plan must describe the same arrangement.
- A ransomware incident runs two clocks at once: the 72-hour reporting clock and your own recovery target. Reports go through the DIBNet portal and need a DoD-approved medium assurance certificate, so check today that someone holds one. Then write down who files the report and who runs the restore. They should not be the same exhausted person.
- Do not wipe and rebuild before you have captured images. The urge to get the network back by Monday can destroy the evidence the clause tells you to keep. Our piece on immutable backups and clean-room recovery covers how to restore into a separate environment while the original systems stay untouched.
Readiness kit
- APC Back-UPS Pro BR1000MS1000 VA / 600 W, sine waveGo to Abt
- EcoFlow DELTA 3 Plus1,024 Wh, under 10 ms UPSGo to EcoFlow
- UGREEN 145W 25000mAh Power Bank25,000 mAh, 145 WGo to UGREEN
Your customer's plan is not your plan
Primes on and around the Arsenal run continuity programs, and they will ask about yours, usually in a supplier questionnaire. The trap is assuming their plan covers you. It covers their mission. If a prime moves work to an alternate site, it may expect your engineers to follow with approved laptops, or it may simply pause your task order until things settle.
Questions to settle with each major customer, in writing where you can:
- During an emergency on the installation, do your on-site staff follow the customer's accountability process, yours, or both? Both is common, and then your list has to be quicker than theirs.
- If gate access is restricted after a storm or a security event, which work can continue off the installation, and on what equipment?
- Which of your deliverables does the customer treat as time critical? Their answer sets your recovery targets better than your own guess. Our primer on RTO and RPO shows how to turn that answer into numbers.
People: clearances are a single point of failure
Small defense IT firms often depend on a handful of cleared staff. If the only engineer with access to a customer's classified network is out sick, on leave, or one of the people whose house was hit, the work stops. You cannot hire a replacement next week either, because a new clearance takes months.
Keep a simple access matrix: each critical customer system down one side, the people with the access and the know-how across the top. Any row with a single name in it is a risk you can see today. Cross-training does not need a clearance for every task. Often the uncleared colleague can handle the documentation, test scripts and customer updates while the cleared engineer works hands-on, and that alone halves the load on one person during an incident.
The same goes for administrative keys: the domain registrar, the email tenant, the firewall, the payroll login. In a small firm those tend to sit with the founder. Put them in a shared password vault with break-glass access for a second named person, and test that access twice a year.
Visibility and cash: the slow risk
The event that ends a small corridor firm is rarely a tornado. It is the recompete that goes to someone else, or a funding pause that stops a task order with thirty days' notice. Continuity in the professional sense covers that too: how many weeks of payroll the business can carry, which contract vehicles you hold, and how a new customer would find you if the old one left.
Firms that keep their capability statement current, keep their SAM.gov registration active and keep their public site describing what they actually do tend to recover faster from a lost contract, because a pipeline already exists. A plan that restores your servers but leaves the business with no next customer has restored the wrong thing.
A one-page version to start
For a firm under fifty people, start with one page and grow it later:
- Who declares an incident, and who is second if that person is unreachable.
- The check-in rule and the channel, with a backup channel for the day the main one fails. The CrowdStrike outage showed how quickly a single tool can take everything with it.
- Where CUI backups live, who can restore them, and who files the report to the Defense Department.
- Customer contacts for each prime, with after-hours numbers.
- The access matrix and the owner of the break-glass vault.
- Power: the generator, or the rule for working from homes that have power, and how many weeks of cash the business can run on.
Test it once a year with a one-hour tabletop. Early spring, before storm season peaks, is a good time; so is the month after a contract award, when the team and the customers change. Our guide to tabletop exercises that expose gaps has scenarios you can adapt, and a north Alabama version almost writes itself: a night warning, a dark office, a ransomware note on Monday morning and a customer asking for status by noon.
The firms that come through those weeks well are rarely the biggest. They are the ones that wrote the plan down while the weather was calm.
Readiness kit
| Category | Pick | Why it is on the list | Link |
|---|---|---|---|
| UPS, sine wave | APC Back-UPS Pro BR1000MS | Sine wave output, 10 outlets and an LCD: a desk UPS for the minutes a desktop needs to save and shut down. | Go to Abt |
| Power station, UPS mode | EcoFlow DELTA 3 Plus | A 1 kWh class station with a UPS mode EcoFlow rates at under 10 ms: the network and a laptop through a full outage day. | Go to EcoFlow |
| Power bank | UGREEN 145W 25000mAh Power Bank | Charges phones and a laptop over USB-C so the station's inverter can stay off. | Go to UGREEN |
Matched to this topic from manufacturer specifications and published reviews; not tested by us. See the Readiness kit



